Skip to yearly menu bar Skip to main content


Workshop Mon, Apr 27, 2026 • 5:00 AM – 1:00 PM PDT 207

Representational Alignment

Badr AlKhamissi ⋅ Brian Cheung ⋅ Dota Tianai Dong ⋅ Stephanie Fu ⋅ Erin Grant ⋅ Kushin Mukherjee ⋅ Ilia Sucholutsky ⋅ SIDDHARTH SURESH
Project Page

Abstract

Video

Chat is not available.

Schedule

Timezone: America/Los_Angeles
5:00 AM
Opening remarks
Badr AlKhamissi ⋅ Erin Grant ⋅ Dota Tianai Dong
6:00 AM
Coffee
7:30 AM
Poster session
9:00 AM
Lunch
11:00 AM
Coffee
12:00 PM
Panel
12:50 PM
Closing remarks
Badr AlKhamissi ⋅ Erin Grant ⋅ Dota Tianai Dong
Training Large Language Models for Self-Explanation Faithfulness
Yeoktatt Cheah ⋅ Maria Perez-Ortiz ⋅ Noah Y Siegel ⋅ Oana-Maria Camburu
Emotion is Not Just a Label: Latent Emotional Factors in LLM Processing
Benjamin Reichman ⋅ Adar Avsian ⋅ Larry Heck
RT-TopKSAE: Improving Top-k Sparse Autoencoders with the Rotation Trick
Sulayman Yusuf ⋅ Saksham Nirvan ⋅ Aishwarya Balwani
Nature vs. Nurture: Limits of Representation Universality Reveal Architectural and Training Differences in Image Models
Fredo Guan ⋅ Shreyan Banerjee ⋅ Aditya Bajoria ⋅ Akarsh Shetty ⋅ Cole Blondin
Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Targeted Sub-Network Updates
Nikhil Prakash ⋅ Donghao Ren ⋅ Dominik Moritz ⋅ Yannick Assogba
Probing and Steering Chain-of-Thought Unfaithfulness in Language Models
Giovanni Occhipinti ⋅ Alessandro Abate ⋅ Nandi Schoots
Does human-alignment benefit interpretability?
Julien Colin ⋅ Nuria Oliver ⋅ Thomas Serre
Enhancing Deep Neural Network Reliability with Refinement and Calibration
Ramya Hebbalaguppe ⋅ ajay shastry ⋅ Soumya Suvra Ghosal ⋅ Chetan Arora
LatentLens: Revealing Highly Interpretable Visual Tokens in LLMs
Benno Krojer ⋅ Shravan Nayak ⋅ Oscar Mañas ⋅ Vaibhav Adlakha ⋅ Desmond Elliott ⋅ Siva Reddy ⋅ Marius Mosbach
Self-supervised EEG pretraining amplifies individual differences in neural representations
Laurine Marty ⋅ Michael Pereira ⋅ Guillaume Bellec ⋅ Joao Barbosa
Do Large Language Models Encode Instruction Overwrites?
You-Chen Lin ⋅ Fu-Chieh Chang ⋅ Pei-Yuan Wu
WHAT TO SHOW AND HOW TO SHOW: STRUCTURED FAILURE-GUIDED SUPERVISION FOR T2I ALIGNMENT
Pulkit Bansal ⋅ Vivek Srivastava ⋅ Shirish Karande
Canonicalizing Multimodal Contrastive Representation Learning
Sharut Gupta ⋅ Sanyam Kansal ⋅ Stefanie Jegelka ⋅ Phillip Isola ⋅ Vikas Garg
From Interaction to Abstraction: Behavior and Neuroimaging Evidence That Reasoning Models Learn Games Like Humans
Botos Csaba ⋅ Sreejan Kumar ⋅ Austin Andrews ⋅ Rui Ponte Costa ⋅ Marcelo G Mattar ⋅ Momchil Tomov
Atlas-Alignment: Making Interpretability Transferable Across Language Models
Bruno Puri ⋅ Jim Berend ⋅ Sebastian Lapuschkin ⋅ Wojciech Samek
Cross-subject decoding of human neural data for speech Brain Computer Interfaces
Tommaso Boccato ⋅ Michał Olak ⋅ Matteo Ferrante
Optimizing Language Models for Crosslingual Knowledge Consistency
Tianyu Liu ⋅ Jirui Qi ⋅ Mrinmaya Sachan ⋅ Ryan Cotterell ⋅ Raquel Fernández ⋅ Arianna Bisazza
OrtSAE: Orthogonal Sparse Autoencoders Uncover Atomic Features
Anton Korznikov ⋅ Andrey Galichin ⋅ Alexey Dontsov ⋅ Oleg Rogov ⋅ Ivan Oseledets ⋅ Elena Tutubalina
Weight Space Detection of Backdoors in LoRA Adapters
David Merenciano ⋅ Ekaterina Vasiagina ⋅ Raghav Dixit ⋅ Kevin Zhu ⋅ Ruizhe Li ⋅ Maheep Chaudhary
Aligning Visual Structural Compositionality in Humans & Vision-Language Models
Helena Balabin ⋅ Lauren DeLong ⋅ Rohan Saha ⋅ Rik Vandenberghe ⋅ Marie-Francine Moens ⋅ Alona Fyshe
THE ILLUSION OF LATENT GENERALIZATION: BI- DIRECTIONALITY AND THE REVERSAL CURSE
Julian Coda-Forno ⋅ Jane Wang ⋅ Arslan Chaudhry
Characterizing Mechanistic Uniqueness and Identifiability Through Circuit Analysis
Ali Sahi ⋅ Imran Kutianawala ⋅ Medhansh Beeram ⋅ Siwoo Song ⋅ Aryan Shrivastava
From Leads to Latents: Attention-Driven Masked Autoencoder for ECG Time Series
Moritz Vandenhirtz ⋅ Samuel Campillo ⋅ Simon Böhi ⋅ Sonia Laguna ⋅ Irene Cannistraci ⋅ Andrea Agostini ⋅ Ece Ozkan ⋅ Thomas Sutter ⋅ Julia E Vogt
Context Sensitivity Improves Human-Machine Visual Alignment
Frieda Born ⋅ Tom Neuhäuser ⋅ Lukas Muttenthaler ⋅ Brett Roads ⋅ Bernhard Spitzer ⋅ Andrew Lampinen ⋅ Matt Jones ⋅ Klaus R Muller ⋅ Michael Mozer
Evolution of CLIP representations: Texture Bias, Human Alignment and Noise Robustness
Pablo Hernández-Cámara ⋅ Jose Jaén-Lorites ⋅ Alex Gomez-Villa ⋅ Jorge Vila Tomás ⋅ Valero Laparra ⋅ Jesus Malo
Task Representational Dynamics For Compositional Generalization During Context-Dependent Cognitive Tasks
Lakshman Chakrav Nallan Chakravarthula ⋅ Michael Cole ⋅ Takuya Ito
IBMA: Information Bottleneck-Based Multimodal Alignment
Yancheng Wang ⋅ Zeyu Dong ⋅ Dongfang Sun ⋅ Alvin Silva ⋅ Teresa Wu ⋅ Yingzhen Yang
Sparks of Rationality: Do Reasoning LLMs Align with Human Judgment and Choice?
Ala N. Tak ⋅ Mohammadamin Banayeeanzade ⋅ Anahita Bolourani ⋅ Fatemeh Bahrani ⋅ Ashutosh Chaubey ⋅ Sai Karimireddy ⋅ Norbert Schwarz ⋅ Jonathan Gratch
TADA! Tuning Audio Diffusion Models through Activation Steering
Łukasz Staniszewski ⋅ Katarzyna Zaleska ⋅ Mateusz Modrzejewski ⋅ Kamil Deja
Steered LLM Activations are Non-Surjective
Aayush Mishra ⋅ Daniel Khashabi ⋅ Anqi Liu
Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs
Pengrui Han ⋅ Xueqiang Xu ⋅ Keyang Xuan ⋅ Peiyang Song ⋅ Siru Ouyang ⋅ Runchu Tian ⋅ Yuqing Jiang ⋅ Cheng Qian ⋅ Pengcheng Jiang ⋅ Jiashuo Sun ⋅ Junxia Cui ⋅ Ming Zhong ⋅ Ge Liu ⋅ Jiawei Han ⋅ Jiaxuan You
Does Downstream Fine-Tuning Undo Embedded Activation Steering?
Philipp E. Glass ⋅ Allan Tucker ⋅ Yongmin Li ⋅ Alina Miron
Seeing the imagined: a latent functional alignment in visual imagery decoding from fMRI data
Fabrizio Spera ⋅ Tommaso Boccato ⋅ Michał Olak ⋅ Sara Cammarota ⋅ Matteo Ciferri ⋅ Michelangelo Tronti ⋅ Nicola Toschi ⋅ Matteo Ferrante
Aligning Human and Mouse EEG representations of Sleep Stages with a Quadruplet Split Latent Permutation Auto-Encoder
Javier Ciudad ⋅ Anders Nørskov ⋅ Alexander Zahid ⋅ Birgitte Kornum ⋅ Morten Mørup
Steering at the Source: Style Modulation Heads for Robust Persona Control
Yoshihiro IZAWA ⋅ Gouki Minegishi ⋅ Koshi Eguchi ⋅ Sosuke Hosokawa ⋅ Kenjiro Taura
Aligning Mass Spectra with Molecular Structure for Fast Computational Olfaction
Ziqi Zhang ⋅ Eunyeong Jin ⋅ Miguel Vasco ⋅ Farzaneh Taleb ⋅ Nona Rajabi ⋅ Alexandra Gutmann ⋅ Jonathan Williams ⋅ Antônio Ribeiro ⋅ Danica Kragic
Verbalizing LLMs' Assumptions to Reduce Sycophancy
Myra Cheng ⋅ Sunny Yu ⋅ Lujain Ibrahim ⋅ Diyi Yang ⋅ Dan Jurafsky
Automated Concept Discovery for LLM-as-a-Judge Preference Analysis
James Wedgwood ⋅ Chhavi Yadav ⋅ Virginia Smith
Information-Theoretic Unsupervised Embedding Quality Evaluation
Mikhail Kuznetsov ⋅ Ivan Butakov ⋅ Marina Munkhoeva ⋅ Alexey Frolov ⋅ Ivan Oseledets
Hair-Trigger Alignment: Black-Box Evaluation Cannot Guarantee Post-Update Alignment
Yavuz Faruk Bakman ⋅ Duygu Nur Yaldiz ⋅ Salman Avestimehr ⋅ Sai Karimireddy
Modulating Cross-Modal Convergence with Single-Stimulus, Intra-Modal Dispersion
Eghbal Hosseini ⋅ Brian Cheung ⋅ Evelina Fedorenko ⋅ Alex Williams
Multi-Way Representation Alignment
Akshit Achara ⋅ Tatiana Gaintseva ⋅ Matéo Mahaut ⋅ Pritish Chakraborty ⋅ Viktor Johansson ⋅ Melih Barsbey ⋅ Emanuele Rodolà ⋅ Donato Crisostomi
Mapping Whisper Representations to Human ECoG Responses with Interpretable Time-Resolved Neural Encoding
Matteo Ciferri ⋅ Tommaso Boccato ⋅ Michał Olak ⋅ Matteo Ferrante ⋅ Nicola Toschi
Align Representations, Not Points: Efficient Rigid Invariant Transport Distance
Zakk Heile ⋅ Peilin He ⋅ Jayson Tran ⋅ Ruiling Wang ⋅ Shrikant Chand
How Data Augmentation Shapes Neural Representations
Tianxiao He ⋅ Alex Williams ⋅ Sarah Harvey
DSPA: Dynamic SAE Steering for Data-Efficient Preference Alignment
James Wedgwood ⋅ Aashiq Muhamed ⋅ Mona Diab ⋅ Virginia Smith
Representational Space Alignment for Vision-Language Models
Tian Yun ⋅ Tarun Kalluri ⋅ Xuefei Cao ⋅ Dat Huynh ⋅ Jiacheng Zhu ⋅ Bor-Chun Chen ⋅ Yifan Wu ⋅ Yuhao Wang ⋅ Ning Zhang ⋅ Miao Liu ⋅ Hengduo Li
Vision-Language Models Fail to Generalize Across Modalities
Yonatan Gideoni ⋅ Yoav Gelberg ⋅ Tim G. J. Rudner ⋅ Yarin Gal
Modeling the language cortex with form-independent and enriched representations of sentence meaning reveals remarkable semantic abstractness
Shreya Saha ⋅ Shurui Li ⋅ Greta Tuckute ⋅ Yuanning Li ⋅ Ru-Yuan Zhang ⋅ Leila Wehbe ⋅ Evelina Fedorenko ⋅ Meenakshi Khosla
Depth-Wise Activation Steering for Honest Language Models
Marysia Winkels ⋅ Gracjan Góral ⋅ Steven Basart